Photoacoustic Image-Guided Delivery of Plasmonic-Nanoparticle

Aug 30, 2018 - In this work, we present a combined ultrasound and photoacoustic imaging tool to provide image-guided needle placement and monitoring o...
0 downloads 0 Views 970KB Size
Subscriber access provided by Kaohsiung Medical University

Communication

Photoacoustic Image Guided Delivery of Plasmonic Nanoparticle Labeled MSCs to the Spinal Cord Eleanor Donnelly, Kelsey P. Kubelick, Diego S. Dumani, and Stanislav Y. Emelianov Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.8b03305 • Publication Date (Web): 30 Aug 2018 Downloaded from http://pubs.acs.org on August 31, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Photoacoustic Image Guided Delivery of Plasmonic Nanoparticle Labeled MSCs to the Spinal Cord Eleanor M. Donnelly1*, Kelsey P. Kubelick2*, Diego S. Dumani1,2, Stanislav Y. Emelianov1,2 1

School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States

2

Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University School of Medicine, Atlanta, Georgia 30332, United States *These authors contributed equally to this work.

KEYWORDS: Stem cells, Photoacoustic imaging, Guided delivery, Ultrasound, Spinal cord

ABSTRACT Regenerative therapies using stem cells have great potential to treat neurodegenerative diseases and traumatic injuries in the spinal cord. In spite of significant research efforts, many therapies fail at the clinical phase. As stem cell technologies advance towards the clinic, there is a need for a minimally invasive, safe, affordable, and real time imaging technique that allows for accurate and safe monitoring of stem cell delivery in real time in the operating room. In this work, we present a combined ultrasound and photoacoustic imaging tool to provide image guided needle placement and monitoring of nanoparticlelabeled stem cell delivery into the spinal cord. We successfully tagged stem cells using gold nanospheres and provided image guided delivery of stem cells into the spinal cord in real time, detecting as few as 1,000 cells. Ultrasound and photoacoustic imaging was used to guide needle placement for direct stem cell injection to minimize risk of needle shear and accidental injury and to improve therapeutic outcomes with accurate, localized stem cell delivery. Following injections of various volumes of cells, three-dimensional ultrasound and photoacoustic images allowed visualization of stem cell distribution along the spinal cord,

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

showing potential to monitor migration of the cells in the future. Feasibility of quantitative imaging was also shown by correlating the total PA signal over the imaging volume to the volume of cells injected. Overall, the presented method may allow clinicians to utilize imaged guided delivery for more accurate and safer stem cell delivery to the spinal cord.

INTRODUCTION Diseases and disorders of the spinal cord are generally met with a poor prognosis, these include neurodegenerative diseases and traumatic injuries of the spinal cord, such as amyotrophic lateral sclerosis (ALS), multiple sclerosis, and spinal cord injury.1 These disorders all include cell death and degeneration of various cell populations within the spinal cord. There are currently many approaches under investigation to treat these neurodegenerative diseases, but one of the most attractive is stem cell transplantation. Stem cell therapy has dual benefits in the setting of the spinal cord by providing cell replacement and trophic support for surviving cells in a neurotoxic environment, such as the inhibitory environment encountered in spinal cord injury. Therapeutic potential of stem cells can be further augmented by genetic modification, for example, to produce therapeutic proteins, such as neurotrophins.2-3 Stem cell therapies have proven to be a highly active research area. Numerous clinical trials utilizing stem cells are under way to treat a plethora of conditions, such as diabetes, osteoarthritis, cancer, and neurological conditions.4-7 Specifically, mesenchymal stem cells (MSCs) are adult-derived multipotent stem cells with the ability to differentiate into many linages, such as chondrocytes, adipocytes, and osteoblasts.8 MSCs have been used in over 800 clinical trials to date, over 40 of these involving spinal cord applications.9 There are three major hurdles to make stem cell transplantation to the spinal cord a more effective therapeutic: 1) accurate real time guidance of stem cell delivery to target areas, 2) noninvasive and longitudinal in vivo cell tracking, and 3) assessment of stem cell viability over time after delivery. In this work we will focus on the first hurdle, real time image guided delivery of stem cells into the spinal cord. Great care and precision is required with delivery of therapeutics into the delicate tissue of the spinal cord,

ACS Paragon Plus Environment

Page 2 of 21

Page 3 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

which is highly sensitive to direct manipulation. Inaccurate needle placement or sudden needle movement can result in damage to the spinal cord white matter tracts, which can have negative outcomes for the patient. Thus, there is a need for a real time, safe, cost effective, and operating room compatible method to provide image guided delivery of cells into the spinal cord. Research has shown that direct accurate injection into the spinal cord parenchyma has a far greater therapeutic benefit than systemic delivery.10 Due to the delicate nature and high risk associated with direct injection into the spinal cord, new methods have been developed to improve safety and efficacy of this procedure.11-12 In one promising approach,12 a custom injection platform is attached to the patient by subcutaneous posts as an alternative to the classic surgical table mounted platform. By fixing the platform to the patient, the needle moves with the patient’s respiration and other movements, which can prevent white matter track damage and inaccurate targeting that can occur with accidental movement. Although development of this device improves the safety of the procedure and the likelihood of accurate stem cell targeting, a blind injection is still required and the procedure could be greatly aided by real time image guidance. From a clinical perspective, implementing real time imaging guidance methods would facilitate more accurate delivery, allowing clinicians to monitor and modify the treatment in real time in the operating room, thus improving likelihood of treatment success. Accurately achieving and confirming correct targeting of the injection is critical to therapeutic outcome.13-14 Classical preoperative imaging, such as magnetic resonance imaging (MRI) can be used to determine injection coordinates a priori. However this is far from ideal. The planned trajectory determined from preoperative imaging may not be feasible in practice during surgery. For example, each patient has unique surface vasculature and vessels must be avoided during injection. Preoperative imaging using MRI does not resolve these vessels, and decisions made by the surgeon to avoid vessels without image-guidance in the operating room may compromise the accuracy of the injection. Overall physicians would greatly benefit from knowing the injection needle was accurately targeted to the correct location of the spinal cord and stem cells were successfully delivered.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Improper needle placement, on the order of millimeters, can be the difference between therapeutic success and failure. Numerous methods have been investigated as potential strategies to label and image transplanted cells in vivo such as bioluminescence, optical imaging, positron emission tomography, and MRI.15-17 While these methods have shown encouraging results, they also possess some disadvantages such as high cost, the need for radiation exposure associated with PET, and difficulty of intraoperative use, making real time feedback challenging. A novel and emerging imaging modality that offers great potential for guiding direct stem cell delivery to the spinal cord and imaging labeled stem cells post-injection is ultrasound (US)-guided photoacoustic (PA) imaging. PA imaging utilizes the photoacoustic effect, which is the conversion of light to sound using a nanosecond pulsed laser. Absorption of electromagnetic radiation by optical absorbers causes localized thermal deposition to produce a pressure wave. As the primary source of contrast comes from optical absorption rather than scattering, highly specific optical contrast is achievable at depths of up to several centimeters. Due to its similarities to US imaging, much of the same hardware, including modified US imaging transducers can be utilized. Furthermore, US imaging provides valuable anatomical information to complement functional information provided by PA imaging. Overall, the familiarity of equipment to clinical staff, small footprint, mobility, lack of radiation exposure, and relatively low cost, make combined ultrasound/photoacoustic (US/PA) imaging an attractive choice for various clinical applications.18-19 There are a wide variety of nanoparticles and dyes which can be utilized as PA imaging contrast agents.20 Plasmonic gold nanospheres (AuNSs) are a promising contrast agent for labeling stem cells for PA image guided delivery.21-23 They have proven to be a very promising photoacoustic contrast agent for labeling and imaging MSCs in vivo.21-22 Prior to transplantation, MSCs are incubated with citrate-stabilized AuNSs. MSCs efficiently uptake these AuNSs without additional manipulation, rendering the cells optically active for PA imaging. AuNSs show no negative effects on cell proliferation, differentiation, or therapeutic potential when tested in vitro.24-26

ACS Paragon Plus Environment

Page 4 of 21

Page 5 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

To date, the utility of US/PA imaging of stem cells in the spinal cord has not yet been investigated. In fact there has been very little use of PA to image the spinal cord in any form.27 In this study, we examined the feasibility of nanoparticles – gold nanospheres (AuNSs), and US/PA imaging as a method to guide direct injection of AuNS-labeled MSCs to an anatomical target in the spinal cord and monitor the injection of labeled stem cells as they are delivered into the spinal cord. We observed a correlation between the volume of cells injected and the magnitude of the PA signal within the imaging volume, which indicates potential for quantitative imaging in real time to determine number of stem cells present. Our results provide proofof-concept that US/PA imaging is a viable option for monitoring stem cell delivery in the spinal cord, with additional advantages compared to other modalities currently under investigation.

RESULTS AND DISCUSSION AuNS synthesis and cell labeling We first produced and characterized AuNSs for use in this study. The absorbance spectrum and TEM of AuNSs were used to confirm successful synthesis. The expected optical absorption peak of 520 nm was observed (Figure 1A-B). We proceeded to label MSCs with AuNSs. MSCs were incubated with AuNSs overnight to allow uptake of the particles. Gold nanoparticles at an optical density (OD) of 0.5 were used for in vitro studies, and an OD of 1 was used to prepare cells for animal studies. An OD of 1 corresponds to approximately 1.5 ng of gold per cells. Unlabeled MSCs were used as a control. As expected based on reports in the literature, the AuNS-labeled MSCs showed a red shift and peak broadening of optical extinction, compared to unlabeled cells (Figure 1C, D). 22 This occurred due to aggregation of the particles within the cells and the resulting plasmon coupling of the AuNSs upon endocytosis.28

The PA properties of the AuNS-labeled MSCs vs naïve (non-labeled) MSCs were assessed using a tissuemimicking gelatin phantom with inclusions containing fixed MSCs. US/PA images were acquired at 700 nm wavelength, and AuNS-labeled MSCs (right) showed a high PA signal relative to the unlabeled control

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(left) (Figure 2A). Further analysis of the PA images showed over an 8-fold increase in the average PA amplitude of AuNS-labeled MSCs compared to unlabeled MSCs at 700 nm (Figure 2B). This high PA signal was obtained from 10,000 cells within each inclusion, well below the amount of cells injected clinically, showing that our approach may be sensitive enough to detect a cell graft in a clinical situation.11 The PA spectrum of the AuNS-labeled MSCs matched the expected absorbance spectrum as previously determined by UV-vis spectrophotometry, with a peak at approximately 700 nm due to surface plasmon resonance coupling (Figure 2C).

Figure 1: UV-Vis characterization of AuNS and MSCs. A; TEM image of AuNSs. B; UV-Vis spectrum of AuNSs in deionized H2O (DiH2O). C; UV-Vis spectrum of fixed unlabeled MSCs. D; UV-Vis spectrum of fixed MSCs labeled with 0.5 OD AuNSs. All cell readings were in a single cell suspension in 1 X PBS, 1 x 106/mL cell concentration.

ACS Paragon Plus Environment

Page 6 of 21

Page 7 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2: Photoacoustic imaging of AuNS-labeled MSCs. A; Ultrasound images (top row, greyscale) and photoacoustic images displayed over the ultrasound images (bottom row, heatmap) of gelatin inclusions containing 1x104 fixed MSCs. Gelatin inclusion on the left contains unlabeled MSCs, inclusion on the right contains MSCs labeled with 0.5 OD AuNSs. The photoacoustic image was obtained at a wavelength of 700 nm. B; Average photoacoustic signal amplitude at 700 nm of gelatin inclusion of fixed unlabeled and fixed AuNS-labeled MSCs. C; Photoacoustic spectrum analysis comparing AuNS-labeled MSCs vs. unlabeled MSCs.

Real time US/PA imaging of needle guidance and cell transplantation in the rodent spinal cord. After proof-of-concept studies visualizing AuNS-labeled MSCs in phantoms, we proceeded to follow the clinical protocol of delivering stem cells into the spinal cord with the aid of US/PA image guidance. First, we wanted to observe that real time US imaging could be used to accurately guide the injection into the spinal cord. The shaft of a 33G steel needle (yellow arrows) was clearly visible (Figure 3A) as the needle tip was continually advanced into the tissue (red arrow) (Figure 3A-D). The supplementary video (Figure SV1) shows needle advancement and targeting to the ventral horn in real time. Results provide proof-of-concept that ultrasound image-guidance can be used to achieve proper needle orientation and depth to reach the ventral portion of the spinal cord (Figure 3D). This real time image guidance would allow surgeons to adjust the injection trajectory in real time, allowing for improved injection accuracy. The angle used for the needle placement was required to visualize the shaft and spinal cord simultaneously, while not compromising the ability to reach the ventral portion of the cord. During the clinical procedure,

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the US/PA transducer will be mounted to the spinal injection platform thus facilitating the translation of US/PA image guided injection by seamless integration with the current injection protocol used clinically.2931

Figure 3: Visualization of needle placement via ultrasound. A-C; Ultrasound images showing the advancement of the 33G steel needle into to ventral portion of the spinal cord. Yellow arrows highlight the needle shaft, red arrow highlights the needle tip. D; Ultrasound image of the final placement of the needle prior to injection of cells into the spinal cord.

Real time guidance is a key advantage of this technology when compared to MRI which is usually performed either preoperatively to obtain target injection coordinates or intraoperatively, neither of which are real time. We chose to target the ventral horn because it is a common therapeutic target for motor neuron diseases. For the treatment of diseases, such as ALS, where a discrete population of cells are affected, accurate and precise delivery of stem cells to the ventral horn is critical for a positive therapeutic outcome. Our results indicate that US/PA imaging offers a potential improvement in the accuracy and precision of injections by giving real time feedback to facilitate immediate adjustments. Real time needle guidance via US/PA imaging also allows for a safer approach. The US/PA transducer can potentially be

ACS Paragon Plus Environment

Page 8 of 21

Page 9 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

coupled with the clinical spinal platform to greatly reduce the risk of damage to white matter tracts during needle placement and during injection. Although needle placement is within the ventral portion of the spinal cord, it is not possible to definitively confirm placement within the gray matter using ultrasound imaging alone. Precise targeting to the gray matter can be achieved by exploiting the different PA properties of the white versus gray matter at 1730 nm.27 Another challenge associated with needle placement is avoiding surface blood vessels, which are oriented in complex patterns. Due to endogenous PA signals from blood, spectroscopic PA imaging could allow visualization of blood vessels during needle guidance, further improving needle targeting. Being able to visualize the blood vessels in real time would allow the surgeon to make real time modifications to the trajectory of the needle, while still accurately reaching the desired target. Following successful placement of the needle, a solution of AuNS-labeled MSCs at 1,000 cells per 1 µL concentration were infused into the rat spinal cord at a rate of 16 nL/sec. A video was continuously acquired throughout the injection of AuNS-labeled MSCs using ultrasound and photoacoustic imaging to visualize injection success (Figure SV2). Frames corresponding to 0, 1, 2, 3, 4, and 5 µL of cell solution injected were further analyzed (Figure 4A-F). At 0 µL, background PA signal from the needle and possibly red blood cells (hemoglobin) within a vessel were visible (Figure 4A). PA signal was observed extending into the tissue from the tip of the needle after only 1 µL of cell suspension was injected, indicating PA imaging was sensitive enough to detect cell numbers as low as 1,000 cells (Figure 4B). US/PA images showed the distribution of cells varied as the infusion proceeded (Figure 4C-F). Qualitatively, the amplitude of the PA signal appeared to decrease at 4 µL and 5 µL, possibly due to cells moving out of the imaging plane (Figure 4E, F). The acquired time course images were further processed using masking and spectroscopic unmixing to separate PA signals from the injection needle, AuNS-labeled MSCs, and endogenous absorbers (Figure 4G). The spectra used for the endogenous absorbers, oxygenated and deoxygenated hemoglobin, were standard within the literature.32 Spectroscopic PA images acquired before the injection were overlaid onto

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the ultrasound images. Oxygenated and deoxygenated hemoglobin (Hb/HbO2) were specifically identified (Figure 4G). Notably, the PA signal at the center of the spinal cord, colored purple, does in fact appear to be from Hb. The PA signal on the top of the cord and in the surrounding muscle, colored green, contained both Hb/HbO2, indicating residual blood from surgery and blood cells remaining in the surrounding vessels. Because the spinal cord moved and deformed slightly upon needle placement and injection, the map of all absorbers does not perfectly align with the PA signals. However, spectroscopic PA image analysis still provided a global map of all absorbers. Results also showed the benefit of implementing multi-wavelength imaging in the imaging protocol to better highlight vasculature and avoid unintentional damage to the spinal cord. Further the ability to distinguish PA signals from stem cells and endogenous absorbers is important for future development of longitudinal tracking. Eosin staining of a transverse section of the spinal cord clearly showed the bolus of transplanted cells within the gray matter, confirming successful delivery to the ventral gray matter of the cord and accuracy of spectroscopic analysis (Figure 4H). The average PA signal amplitude was analyzed over 10 frames at a 2-dimensional transverse cross-section for each injection volume (Figure 4I). An approximately 2-fold increase in average PA signal was observed from 0 µl to 1 µL of cells injected. However, the average PA signal quickly plateaus and even decreased slightly as the injection proceeded up to 5 µL. The decreased average PA signal in the time course images may have been due to a saturation of PA signal; however, it was more likely that cells moved horizontally out of the imaging plane. Overall results indicate that AuNSs or similar photoacoustic imaging contrast agent are a feasible option for evaluating successful stem cell injection and later track transplanted stem cells within the spinal cord longitudinally.

Quantitative assessment of PA imaging of AuNS-labeled stem cells. In order to further assess the utility of PA imaging to image AuNS-labeled MSCs, we performed a dose escalation study to evaluate potential for quantitative imaging. We chose a range of cell numbers from 1 x 104 – 4 x 104 cells, at a density of 1 x 104 cells/µL and injected along the length of the spinal cord. We chose

ACS Paragon Plus Environment

Page 10 of 21

Page 11 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

1, 2, and 4 µL injections volumes. After the injection, the spinal cord was imaged using photoacoustic tomography in order to obtain volumetric three-dimensional (3D) data. The difference in volume of the stem cell graft can be seen in Figure 5A. Overall, the PA signal appears to increase relative to cell volume injected. The PA amplitude corresponding to each injection volume was analyzed and plotted as seen in Figure 5B. There was clearly a linear relationship between volume of cells injected vs the total PA signal obtained. This highlights the utility of PA imaging of labeled cells as a quantitative tool. This quantitative capability has two major advantages. Firstly, in the operating room, real-time quantitative imaging can allow the actual cell number delivered to be ascertained, and therefore additional cells may be delivered without the need for an additional surgery. Secondly, when the US/PA imaging approach advances to in vivo cell tracking, quantitative imaging will allow determination of the number of cells remaining at the injection site and the number of migrated cells. This information would improve understanding of stem cell behavior in the spinal cord in vivo. A more complete study needs to be performed to further assess the full quantitative capabilities of PA imaging of stem cells and to construct a suitable standard curve for relating cell number to PA signal.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

 Figure 4: US/PA imaging of injection of AuNS-labeled MSCs into the rodent spinal cord. A-F; Ultrasound and photoacoustic images of injection of AuNS-labeled MSCs into the rodent spinal cord, taken at 0, 1, 2, 3, 4, and 5 µL injection volume, respectively. Scale bar is 2 mm. G; Distribution of all photoacoustic absorbers after 5 µL injection of AuNS-labeled MCSs. Spectroscopic photoacoustic image processing followed by image segmentation and masking to distinguish PA signals from the needle, AuNS-labeled MSCs, and endogenous absorbers. Red indicates the needle, yellow indicates AuNS-labeled MSCs, blue indicates oxygenated hemoglobin, purple indicates deoxygenated hemoglobin, and green indicates both species of hemoglobin. This representative map clearly separates AuNS-labeled MSCs from background absorbers. H; Representative photomicrograph showing transplanted AuNS-labeled MSCs in the gray matter of the spinal cord. Scale bar is 2 mm. I; Average amplitude of photoacoustic signal at 700 nm during injection. Average of ten frames plotted for each volume point, error bars indicate standard deviation.

Figure 5: Quantitative study. A: Reconstructed 3D photoacoustic image of the spinal cord with injected cells denoted by arrows. B; Photoacoustic signal from the injection of increasing volumes of AuNS-labeled MSCs into rodent spinal cord.

The next steps in this work will involve imaging labeled MSCs in the spinal cord in vivo longitudinally to observe their viability, migration, and engraftment over an acute period. Also analysis of the quantitative

ACS Paragon Plus Environment

Page 12 of 21

Page 13 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

capabilities of this approach would further increase its utility. Following this, US/PA imaging of stem cells in the spinal cord could be investigated for translation to current clinical approaches to treat ALS and spinal cord injury.

CONCLUSIONS Stem cells are fast advancing into the clinic for a plethora of applications. Despite great advances in stem cell treatments, there are still numerous gaps in the knowledge base that need to be filled before stem cells can reach their full clinical potential. The ability to achieve precise, real time image guided delivery of stem cells would be a highly valuable tool in research and in the clinic. In this work, we presented a photoacoustic contrast agent for imaging stem cells within the rodent spinal cord. We successfully labeled human MSCs with AuNSs and were able to image these labeled cells in real time while being injected into the rat spinal cord, using ultrasound and photoacoustic imaging. The labeled MSCs showed high photoacoustic signals above background. We clearly visualized signal from as few as 1,000 cells, which is below the cell number typically used clinically. We also demonstrated the capabilities of photoacoustic imaging to provide quantitative information on the volume of cells injected. We believe that photoacoustic and ultrasound imaging, in combination with AuNSs or other imaging contrast agents, is a viable alternative to currently available imaging modalities and can allow for real time in vivo imaging of stem cell delivery into the spinal cord. This platform will provide clinicians and researchers valuable information to advance the field of stem cell therapies. US/PA imaging can provide safe, accurate, real time, and operating room compatible guidance of stem cell delivery into the spinal cord.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

METHODS Synthesis and characterization of gold nanospheres (AuNSs) Gold nanospheres (AuNSs) were synthesized as described elsewhere,21 and the optical absorption spectrum and particle morphology were assessed to confirm successful synthesis. See Supporting Information for further details. Labeling of mesenchymal stem cells (MSCs) with AuNSs MSCs were incubated with sterilized AuNSs at a concentration of 0.5 optical density (OD) for 24 hrs. The media containing AuNSs was aspirated. Remaining gold nanosphere-labeled MSCs (AuNS-labeled MSCs) were harvested by trypsinization and resuspended in 1 X PBS at the desired concentration. See Supporting Information for further details. US/PA imaging of AuNS-labeled MSCs in vitro Fixed cells at a concentration of 10,000 cells per 1 µL were used for tissue-mimicking phantom experiments to assess successful labeling of MSCs for PA imaging. For a detailed recipe of the tissue-mimicking gelatin phantom and cell inclusions, see Supporting Information. The gelatin phantom with inclusions was imaged using the Vevo LAZR (Fujifilm VisualSonics Inc., Toronto, Canada) – a combined US/PA imaging system. Details on the Vevo LAZR hardware and imaging parameters are described in Supporting Information. The generated data was exported and postprocessed in MATLAB (MathWorks, Inc., Natick, MA). Ex vivo imaging of AuNS-labeled MSCs in rodent spinal cord All experiments involving animals were performed under the guidelines of the Institutional Animal Care and Use Committee (IACUC) of the Georgia Institute of Technology. Female Sprague Dawley rats were euthanized by CO2 asphyxiation followed by cervical dislocation. The lumbar ACS Paragon Plus Environment

Page 14 of 21

Page 15 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

spinal column was removed and a multilevel lumbar laminectomy was performed to expose the spinal cord. The ventral portion of the spinal cord was injected with a 5 µl volume containing 5,000 unfixed AuNS-labeled MSCs in 1 X PBS at a rate of 16 nL/sec using a 33G nanofil syringe attached to an ultra-micropump (WPI, Sarasota, FL). US/PA imaging to guide AuNS-labeled MSC injection and confirm delivery was performed in multi- and single-wavelength imaging modes using the Vevo LAZR system. See Supporting Information for details on the imaging protocol. Spectroscopic processing of multi-wavelength PA images was based on previously methods33 described elsewhere (see also Supporting Information). Quantitative 3-D photoacoustic imaging of cell injection volumes For the quantitative study, three injections of AuNS-labeled MSCs were performed at different locations along the spinal cord, similarly using an ultra-micropump. The injection volumes were 1, 2, and 4 µL. The concentration of all injections was 10,000 cells/µL. After all injections, singlewavelength,

volumetric

three-dimensional

PA

imaging

was

performed

using

a

photoacoustic/fluorescent tomography (PAFT) system (PhotoSound Technologies Inc., Houston, TX). The PAFT system is described in detail in Supporting Information. Tomographic singlewavelength PA datasets were exported for post-processing. The three-dimensional tomographic PA image was displayed using AMIRA (Thermo Scientific, Waltham, MA). Each injection was visually distinguished based on magnitude of the PA signal at 700 nm. Thus, red pixels primarily corresponded to locations of AuNS-labeled MSC injections. Total photoacoustic signal intensity for 1, 2, and 4 µL injections was calculated using MATLAB. The reconstructed three-dimensional image was thresholded to remove the noise floor and was integrated along the coronal plane to produce a two-dimensional sagittal projection image. A smaller ROI was selected from the two-dimensional image to remove signals from non spinal cord

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

tissue such as bone and muscle. It was then assumed that any remaining PA signals in the ROI primarily come from AuNS-labeled MSCs. The sagittal image was integrated again to produce a one-dimensional plot of PA signal intensity along the spinal cord (rostral to caudal). Three peaks were found, corresponding to the 1 µL, 2 µL, or 4 µL injections. The half-maximum was calculated for each peak and used to further threshold the images. Slices with PA signal intensities above the half-maximum threshold were used to calculate the total PA signal for each injection. The linear best fit was calculated to demonstrate the dependence of total PA signal with injected volume of cells. See Supporting Information for additional details on experimental procedures and processing used for tomographic quantitative analysis.   

ACS Paragon Plus Environment

Page 16 of 21

Page 17 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

SUPPORTING INFORMATION Detailed methods (PDF) Real-time B-Mode ultrasound video of imaged-guided placement of the needle into the ventral horn of the rodent spinal cord (AVI) US/PA video captured as 3 µL of AuNS-labeled MCSs were injected into the rodent spinal cord. The video is at 10x speed (AVI)

Corresponding Author ‡ Email: [email protected]

ORCID Kelsey P. Kubelick: 0000-0002-6342-2994 Diego S. Dumani: 0000-0002-0086-4250 Stanislav Emelianov: 0000-0002-7098-133X

Author Contributions E.M.D. conceived the idea. E.M.D., K.P.K. and S.Y.E. discussed and planned the studies. E.M.D. prepared samples and performed ex vivo surgeries. K.P.K., D.S.D. and E.M.D. performed imaging and data analysis. E.M.D., K.P.K., D.S.D. and S.Y.E. prepared and edited the manuscript.

Funding Sources This work was supported in part by National Institutes of Health under grants NS102860 and EB015007.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACKNOWLEDGMENTS We wish to acknowledge Nicholas Boulis, MD, Professor of Neurosurgery at Emory University School of Medicine, for his critical feedback and insightful discussion. This work was performed in part at Georgia Tech’s Institute for Electronics and Nanotechnology, a member of the National Nanotechnology Coordinated Infrastructure, which is supported by the National Science Foundation (Grant ECCS-1542174).

 

ACS Paragon Plus Environment

Page 18 of 21

Page 19 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

References 1. 2.

3.

4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

17.

Donnelly, E. M.; Lamanna, J.; Boulis, N. M., Stem cell therapy for the spinal cord. Stem Cell Res Ther 2012, 3 (4), 24. Stewart, A. N.; Kendziorski, G.; Deak, Z. M.; Bartosek, N. C.; Rezmer, B. E.; Jenrow, K.; Rossignol, J.; Dunbar, G. L., Transplantation of Mesenchymal Stem Cells that Overexpress NT-3 Produce Motor Improvements without Axonal Regeneration following Complete Spinal Cord Transections in Rats. Brain Res 2018. Thomsen, G. M.; Avalos, P.; Ma, A. A.; Alkaslasi, M.; Cho, N.; Wyss, L.; Vit, J. P.; Godoy, M.; Suezaki, P.; Shelest, O.; Bankiewicz, K. S.; Svendsen, C. N., Transplantation of Neural Progenitor Cells Expressing Glial Cell Line-Derived Neurotrophic Factor into the Motor Cortex as a Strategy to Treat Amyotrophic Lateral Sclerosis. Stem Cells 2018. Moreira, A.; Kahlenberg, S.; Hornsby, P., Therapeutic potential of mesenchymal stem cells for diabetes. J Mol Endocrinol 2017, 59 (3), R109-R120. Berebichez-Fridman, R.; Gomez-Garcia, R.; Granados-Montiel, J.; Berebichez-Fastlicht, E.; OlivosMeza, A.; Granados, J.; Velasquillo, C.; Ibarra, C., The Holy Grail of Orthopedic Surgery: Mesenchymal Stem Cells-Their Current Uses and Potential Applications. Stem Cells Int 2017, 2017, 2638305. Zhang, C. L.; Huang, T.; Wu, B. L.; He, W. X.; Liu, D., Stem cells in cancer therapy: opportunities and challenges. Oncotarget 2017, 8 (43), 75756-75766. Lo Furno, D.; Mannino, G.; Giuffrida, R., Functional role of mesenchymal stem cells in the treatment of chronic neurodegenerative diseases. J Cell Physiol 2017. Samsonraj, R. M.; Raghunath, M.; Nurcombe, V.; Hui, J. H.; van Wijnen, A. J.; Cool, S. M., Concise Review: Multifaceted Characterization of Human Mesenchymal Stem Cells for Use in Regenerative Medicine. Stem Cells Transl Med 2017. https://clinicaltrials.gov/. https://clinicaltrials.gov/. Lamanna, J. J.; Miller, J. H.; Riley, J. P.; Hurtig, C. V.; Boulis, N. M., Cellular therapeutics delivery to the spinal cord: technical considerations for clinical application. Ther Deliv 2013, 4 (11), 1397-410. Feldman, E. L.; Boulis, N. M.; Hur, J.; Johe, K.; Rutkove, S. B.; Federici, T.; Polak, M.; Bordeau, J.; Sakowski, S. A.; Glass, J. D., Intraspinal neural stem cell transplantation in amyotrophic lateral sclerosis: phase 1 trial outcomes. Ann Neurol 2014, 75 (3), 363-73. Riley, J.; Glass, J.; Feldman, E. L.; Polak, M.; Bordeau, J.; Federici, T.; Johe, K.; Boulis, N. M., Intraspinal stem cell transplantation in amyotrophic lateral sclerosis: a phase I trial, cervical microinjection, and final surgical safety outcomes. Neurosurgery 2014, 74 (1), 77-87. Bartus, R. T.; Johnson, E. M., Jr., Clinical tests of neurotrophic factors for human neurodegenerative diseases, part 1: Where have we been and what have we learned? Neurobiol Dis 2017, 97 (Pt B), 156168. Bartus, R. T.; Johnson, E. M., Jr., Clinical tests of neurotrophic factors for human neurodegenerative diseases, part 2: Where do we stand and where must we go next? Neurobiol Dis 2017, 97 (Pt B), 169178. Sun, N.; Lee, A.; Wu, J. C., Long term non-invasive imaging of embryonic stem cells using reporter genes. Nat Protoc 2009, 4 (8), 1192-201. Boddington, S. E.; Henning, T. D.; Jha, P.; Schlieve, C. R.; Mandrussow, L.; DeNardo, D.; Bernstein, H. S.; Ritner, C.; Golovko, D.; Lu, Y.; Zhao, S.; Daldrup-Link, H. E., Labeling human embryonic stem cell-derived cardiomyocytes with indocyanine green for noninvasive tracking with optical imaging: an FDA-compatible alternative to firefly luciferase. Cell Transplant 2010, 19 (1), 55-65. Lamanna, J. J.; Gutierrez, J.; Urquia, L. N.; Hurtig, C. V.; Amador, E.; Grin, N.; Svendsen, C. N.; Federici, T.; Oshinski, J. N.; Boulis, N. M., Ferumoxytol Labeling of Human Neural Progenitor Cells for Diagnostic Cellular Tracking in the Porcine Spinal Cord with Magnetic Resonance Imaging. Stem Cells Transl Med 2017, 6 (1), 139-150.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

18. Emelianov, S. Y.; Li, P. C.; O'Donnell, M., Photoacoustics for molecular imaging and therapy. Phys Today 2009, 62 (8), 34-39. 19. Mallidi, S.; Luke, G. P.; Emelianov, S., Photoacoustic imaging in cancer detection, diagnosis, and treatment guidance. Trends Biotechnol 2011, 29 (5), 213-21. 20. Luke, G. P.; Yeager, D.; Emelianov, S. Y., Biomedical applications of photoacoustic imaging with exogenous contrast agents. Ann Biomed Eng 2012, 40 (2), 422-37. 21. Ricles, L. M.; Nam, S. Y.; Sokolov, K.; Emelianov, S. Y.; Suggs, L. J., Function of mesenchymal stem cells following loading of gold nanotracers. Int J Nanomedicine 2011, 6, 407-16. 22. Nam, S. Y.; Ricles, L. M.; Suggs, L. J.; Emelianov, S. Y., In vivo ultrasound and photoacoustic monitoring of mesenchymal stem cells labeled with gold nanotracers. PLoS One 2012, 7 (5), e37267. 23. Chung, E.; Nam, S. Y.; Ricles, L. M.; Emelianov, S. Y.; Suggs, L. J., Evaluation of gold nanotracers to track adipose-derived stem cells in a PEGylated fibrin gel for dermal tissue engineering applications. Int J Nanomedicine 2013, 8, 325-36. 24. Betzer, O.; Meir, R.; Dreifuss, T.; Shamalov, K.; Motiei, M.; Shwartz, A.; Baranes, K.; Cohen, C. J.; Shraga-Heled, N.; Ofir, R.; Yadid, G.; Popovtzer, R., In-vitro Optimization of Nanoparticle-Cell Labeling Protocols for In-vivo Cell Tracking Applications. Sci Rep 2015, 5, 15400. 25. Wan, D.; Chen, D.; Li, K.; Qu, Y.; Sun, K.; Tao, K.; Dai, K.; Ai, S., Gold Nanoparticles as a Potential Cellular Probe for Tracking of Stem Cells in Bone Regeneration Using Dual-Energy Computed Tomography. ACS Appl Mater Interfaces 2016, 8 (47), 32241-32249. 26. Nold, P.; Hartmann, R.; Feliu, N.; Kantner, K.; Gamal, M.; Pelaz, B.; Huhn, J.; Sun, X.; Jungebluth, P.; Del Pino, P.; Hackstein, H.; Macchiarini, P.; Parak, W. J.; Brendel, C., Optimizing conditions for labeling of mesenchymal stromal cells (MSCs) with gold nanoparticles: a prerequisite for in vivo tracking of MSCs. J Nanobiotechnology 2017, 15 (1), 24. 27. Wu, W.; Wang, P.; Cheng, J. X.; Xu, X. M., Assessment of white matter loss using bond-selective photoacoustic imaging in a rat model of contusive spinal cord injury. J Neurotrauma 2014, 31 (24), 1998-2002. 28. Mallidi, S.; Larson, T.; Aaron, J.; Sokolov, K.; Emelianov, S., Molecular specific optoacoustic imaging with plasmonic nanoparticles. Opt Express 2007, 15 (11), 6583-8. 29. Riley, J.; Federici, T.; Park, J.; Suzuki, M.; Franz, C. K.; Tork, C.; McHugh, J.; Teng, Q.; Svendsen, C.; Boulis, N. M., Cervical spinal cord therapeutics delivery: preclinical safety validation of a stabilized microinjection platform. Neurosurgery 2009, 65 (4), 754-61; discussion 761-2. 30. Glass, J. D.; Hertzberg, V. S.; Boulis, N. M.; Riley, J.; Federici, T.; Polak, M.; Bordeau, J.; Fournier, C.; Johe, K.; Hazel, T.; Cudkowicz, M.; Atassi, N.; Borges, L. F.; Rutkove, S. B.; Duell, J.; Patil, P. G.; Goutman, S. A.; Feldman, E. L., Transplantation of spinal cord-derived neural stem cells for ALS: Analysis of phase 1 and 2 trials. Neurology 2016, 87 (4), 392-400. 31. Riley, J.; Butler, J.; Baker, K. B.; McClelland, S., 3rd; Teng, Q.; Yang, J.; Garrity-Moses, M.; Federici, T.; Boulis, N. M., Targeted spinal cord therapeutics delivery: stabilized platform and microelectrode recording guidance validation. Stereotact Funct Neurosurg 2008, 86 (2), 67-74. 32. https://omlc.org/spectra/hemoglobin/index.html. 33. Kim, S.; Chen, Y. S.; Luke, G. P.; Emelianov, S. Y., In vivo three-dimensional spectroscopic photoacoustic imaging for monitoring nanoparticle delivery. Biomed Opt Express 2011, 2 (9), 254050.

ACS Paragon Plus Environment

Page 20 of 21

Page 21 of 21

+

0 .8

=

L a b e le d M S C s U n la b e le d M S C s

Needle Spinal cord

0 .6

0 .4

0 .2

Unlabeled MSCs

0 .0 700

800

Needle 350

AuNS-labeled MSCs

900

Ultrasound/ photoacoustic

amplitude (a.u.)

1 .0

Image guided delivery of stem cells to the spinal cord

AuNS-labeled MSC

Ultrasound

Mesenchymal Gold nanospheres stem cell (MSC) (AuNSs)

P APhotoacoustic s ig n a l a m p litusignal d e ( a .u .)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Nano Letters

400

450

500

WWavelength a v e le n g th (n(nm) m)

ACS Paragon Plus Environment

AuNS-labeled MSCs 50

100

150

200

250

300

350